Modeling subgrid-scale scalar dissipation rate in turbulent premixed flames using gene expression programming and deep artificial neural networks

Kasten, Christian; Shin, Junsu; Sandberg, Richard D.; Pfitzner, Michael; Chakraborty, Nilanjan; Klein, Markus

doi:10.1063/5.0095886

Cited by 10 publications

(5 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, it is very important that any evolutionary algorithm used in production should be able to find as good solutions as possible in real engineering problems as well. Preliminary tests in the field of turbulent combustion modeling [37], [38], not shown here, indicate that GGEP provides models that are at least 20% better error wise than those from GEP, which can be considered a strong indication that GGEP is ready to be used in productive applications. The question remains why GGEP reaches such high levels of performance and effectiveness.…”

Section: Discussionmentioning

confidence: 73%

See 1 more Smart Citation

An Efficient Way of Introducing Gender Into Evolutionary Algorithms

Kasten

Fahr

Klein

2023

IEEE Trans. Evol. Computat.

View full text Add to dashboard Cite

Evolutionary algorithms have been extensively used for numerous optimization problems with great success in the past years. They mimic nature's process of evolution to find solutions to a large range of mathematical problems. In this work a new strategy is suggested to introduce gender, defined by a characteristic of an individual, that is easy to implement in evolutionary algorithms, as long as they are based on evaluating a fitness function. The new method outperforms comparable evolutionary approaches without gender for all standard test problems considered. The present study shows that with increasing problem complexity the performance of this gender variant increases to more than double the success rates while keeping the computational effort at about the same level and still being clearly better for easy problems. Additionally, in the mean, the new method results in better fitness values for all presented cases.

show abstract

Section: Discussionmentioning

confidence: 73%

“…Here, the dominance of GGEP in regards of the first three factors will be demonstrated in detail using problems 1 to 3. The fourth criterion will be relevant for much more complex optimization problems, such as turbulent heat transfer or combustion modeling in the field of computational fluid dynamics [37], [38].…”

Section: A Test Problem Descriptionmentioning

confidence: 99%

An Efficient Way of Introducing Gender Into Evolutionary Algorithms

Kasten

Fahr

Klein

2023

IEEE Trans. Evol. Computat.

View full text Add to dashboard Cite

show abstract

“…ML has also been used to predict the subgrid micromixing rate, employing techniques like gene expression programming. 362 A physics-informed enhanced super-resolution Generative adversarial network (GAN) has been employed for subfilter modeling of mixing. 363 Physics-informed constraints are integrated into the loss function to guide the reconstruction process, enhancing model generalization and interpretability.…”

Section: Macroscopic Mixtures Modelingmentioning

confidence: 99%

“…ML has also been used to predict the subgrid micromixing rate, employing techniques like gene expression programming . A physics-informed enhanced super-resolution Generative adversarial network (GAN) has been employed for subfilter modeling of mixing .…”

Section: Continuum-scale Chemical Mixturesmentioning

confidence: 99%

Mixtures Recomposition by Neural Nets: A Multidisciplinary Overview

Nicolle,

Deng,

Ihme

et al. 2024

J. Chem. Inf. Model.

View full text Add to dashboard Cite

Artificial Neural Networks (ANNs) are transforming how we understand chemical mixtures, providing an expressive view of the chemical space and multiscale processes. Their hybridization with physical knowledge can bridge the gap between predictivity and understanding of the underlying processes. This overview explores recent progress in ANNs, particularly their potential in the ’recomposition’ of chemical mixtures. Graph-based representations reveal patterns among mixture components, and deep learning models excel in capturing complexity and symmetries when compared to traditional Quantitative Structure–Property Relationship models. Key components, such as Hamiltonian networks and convolution operations, play a central role in representing multiscale mixtures. The integration of ANNs with Chemical Reaction Networks and Physics-Informed Neural Networks for inverse chemical kinetic problems is also examined. The combination of sensors with ANNs shows promise in optical and biomimetic applications. A common ground is identified in the context of statistical physics, where ANN-based methods iteratively adapt their models by blending their initial states with training data. The concept of mixture recomposition unveils a reciprocal inspiration between ANNs and reactive mixtures, highlighting learning behaviors influenced by the training environment.

show abstract

“…Computational fluid dynamics (CFD) is not exempt. ML tools were applied in modeling [1][2][3][4][5][6][7][8], computation [9][10][11], control [12], and optimization [13][14][15]. An overview of the ML applications in the field of fluid dynamics can be found in Refs.…”

Section: Introductionmentioning

confidence: 99%

A survey of machine learning wall models for large eddy simulation

Vadrot¹,

Yang²,

Abkar³

2022

Preprint

View full text Add to dashboard Cite

This survey investigates wall modeling in large eddy simulations (LES) using data-driven machine learning (ML) techniques. To this end, we implement three ML wall models in an open-source code and compare their performances with the equilibrium wall model in LES of half-channel flow at eleven friction Reynolds numbers between 180 and 10 10 . The three models have "seen" flows at only a few Reynolds numbers. We test if these ML wall models can extrapolate to unseen Reynolds numbers. Among the three models, two are supervised ML models, and one is a reinforcement learning ML model. The two supervised ML models are trained against direct numerical simulation (DNS) data, whereas the reinforcement learning ML model is trained in the context of a wallmodeled LES with no access to high-fidelity data. The two supervised ML models capture the law of the wall at both seen and unseen Reynolds numbers-although one model requires re-training and predicts a smaller von Kármán constant. The reinforcement learning model captures the law of the wall reasonably well but has errors at both low (Reτ < 10 3 ) and high Reynolds numbers (Reτ > 10 6 ). In addition to documenting the results, we try to "understand" why the ML models behave the way they behave. Analysis shows that the errors of the supervised ML model is a result of the network design and the errors in the reinforcement learning model arise due to the present choice of the "states" and the mismatch between the neutral line and the line separating the action map. In all, we see promises in data-driven machine learning models.I.

show abstract

Modeling subgrid-scale scalar dissipation rate in turbulent premixed flames using gene expression programming and deep artificial neural networks

Cited by 10 publications

References 53 publications

An Efficient Way of Introducing Gender Into Evolutionary Algorithms

An Efficient Way of Introducing Gender Into Evolutionary Algorithms

Mixtures Recomposition by Neural Nets: A Multidisciplinary Overview

A survey of machine learning wall models for large eddy simulation

Contact Info

Product

Resources

About